I summarize here two projects in which we have made significant progress during the fiscal year 2008. [unreadable] [unreadable] Neuronal projection from the posterior hypothalamus, particularly the supramammillary nucleus, to the septum may be involved in reward. Our group recently found that rats learn to self-administer several chemicals into the supramammillary nucleus: nicotine; GABAA receptor antagonists; and the excitatory amino acid AMPA. Because the supramammillary neurons send their axons to the septum and because those neurons contain glutamate, Rick Shin and I examined the hypothesis that AMPA administration into the septum is rewarding, using intracranial self-administration procedures. Rats (N = 62) received AMPA injections (100 nl volume of 0.01, 0.05, and 0.25 mM) for various septal regions via permanent guide cannulae. Vigorous self-administration of AMPA was found when AMPA was administered along the midline, particularly into the medial septum. Rats (N = 11) self-administer 0.05 or 0.25 mM AMPA into the medial septum at mean rates of 0.9 infusions per minute in 90 minute sessions. Administration of AMPA into the lateral septum or medial accumbens shell was less effective than those into the medial septum. In addition, when AMPA was co-infused into the medial septum with the competitive AMPA receptor antagonist ZK 200775 (0.1 mM), self-administration was diminished. These data suggest that the stimulation of AMPA receptors in the medial septum is rewarding. Glutamatergic innervation from the supramammillary nucleus to the medial septum may be involved in reward. [unreadable] [unreadable] The other project concerns our observation that energized approach by amphetamine administration into the ventral striatum depends on the dorsomedial striatum. We have recently found that intermittent, non-contingent injections of amphetamine into the ventral striatum, especially the medial olfactory tubercle (mOT), energize lever-pressing reinforced by cue light. Because anatomical data suggest that signals processed in the mOT eventually were sent to the dorsal striatum, we examined such energizing effects of amphetamine triggered from the mOT are mediated by serial connectivity with the dorsomedial striatum (DMS). Each rat received the permanent implantation of two guide cannulae, one targeting the mOT and the other just above the ipsilateral or contralateral DMS. In each session, the rat received 60 injections into the mOT with a 90-sec fixed-interval schedule over 90 min in the test chamber equipped with two levers. Responding on one lever turned on cue light for one second, whereas responding on the other lever had no programmed consequence. Animals received mOT infusions of artificial cerebrospinal fluid (ACSF) in session 1 and 30 mM amphetamine thereafter. In sessions 1-3 and 6-7, rats received no DMS injections, whereas in sessions 4 and 5, an injection cannula (31 gauge) was inserted, extending 3 mm below the tip of the guide, into the DMS and ACSF (0.5 l) was infused over 30 sec just before behavioral testing. Consistent with our previous findings, amphetamine infusions into the mOT increased lever presses in sessions 2 and 3. Lever-press levels in these sessions did not differ between rats with ipsilateral and contralateral DMS guides. However, mere insertion of the injection cannula and ACSF infusion into the ipsilateral, but not contralateral, DMS just prior to sessions 4 and 5 prevented mOT amphetamine from increasing lever-presses (F1,15=4.77, P<0.05). The rats with the ipsilateral guide restored lever-press levels in sessions 6 and 7 when they did not receive the DMS treatment. While additional experiments are ongoing, these data tentatively suggest that the DMS plays an important role in expression of approach facilitated by enhanced dopamine transmission in the mOT. The DMS may receive motivational signals from the ventromedial striatum (mOT) via the ventral pallidum-to-dorsomedial thalamus-to-prefrontal cortex circuitry and engage in serial processes with these regions for motivated behavior.